China

Researchers propose DPMS-Net, a deep network that uses dynamic convolution to mine discriminative cues across temporal, spatial and frequency dimensions, combines channel and temporal attention, and adds a spectral-domain analysis component to surface subtle oscillatory features hidden in the EEG spectrum. On the BCI Competition IV 2a and 2b datasets it reached subject-dependent accuracies of 83.93% and 88.38%, and 67.67% on a self-collected stroke-patient dataset. The authors say its efficient decoding and robustness suit neurorehabilitation BCI systems.

The study presents MAPS-CS, an online SSVEP brain-computer interface that selects channels dynamically during the experiment. A multi-dimensional feature framework covering signal energy, stability and inter-channel correlation quantifies anomalies and generates scores that a hierarchical decision step combines into a channel-quality score to identify and remove bad channels — with no training required. Against the channel ensemble (CE) method, MAPS-CS lifted standard FBCCA accuracy by 3.5%, 4.1%, 4.4% and 6.5% at stimulus durations of 2 s, 1.5 s, 1 s and 0.5 s, the best among the CE, binary harmony search and TOP-K local optimization methods compared.

A randomized controlled trial in 120 patients with post-stroke hemiplegia found that pairing brain-computer interface (BCI) rehabilitation training with a gait robot accelerated lower-limb recovery and improved muscle activation patterns more than either intervention on its own. After eight weeks, BCI training, gait-robot training and the combination all improved lower-limb motor function, muscle activity and gait compared with routine training, but the combined group showed the largest increases in integrated electromyography of the tibialis anterior and gastrocnemius, along with a greater gain in step frequency and a greater reduction in step width.

A new entry in the Chinese Clinical Trial Registry covers a study using brain-computer interface technology for functional recovery after stroke. The record, ChiCTR2600120924, was posted on March 23, 2026, and lists its status as not yet recruiting. The registry listing carries only bibliographic details, so the trial design, sample size and sponsor remain undisclosed.

Nine Jiangsu provincial departments in eastern China led by the Department of Industry and Information Technology issued the Action Plan for Innovative Development of the Brain-Computer Interface Industry in Jiangsu in March 2026, implementing national guidance from seven ministries. Its most distinctive provision covers industrial deployment, directing firms in hazardous materials, nuclear power, mining and electricity to pilot BCI systems in safety management and deep-sea and deep-pit operations, sectors no other provincial document names. Headline goals are at least two provincial clusters by 2027 and two to three leading enterprises by 2030. Its quantitative targets sit at the end of 14 tasks and all fall in 2030, including at least 15 programme projects, 15 key technology tasks, about 10 innovation platforms, 30 certified flagship products, 20 registered medical devices and a dataset of at least 20,000 samples.

A Chinese trial will measure the immediate effects of a non-invasive brain-machine interface combined with multimodal training on neurophysiological function in patients with upper-limb hemiparesis after stroke. It was registered with the Chinese Clinical Trial Registry as ChiCTR2600120046 on March 9, 2026 and has not begun recruiting.

Four Chinese ministries jointly issued science-technology insurance guidance, listing BCI alongside AI and quantum technology as key sectors for dedicated insurance products.

Researchers have proposed MSARFNet, a multi-scale attention-based reconstruction fusion network that reached average classification accuracies of 84.64% and 87.96% on the BCI Competition IV 2a and 2b motor imagery datasets, outperforming several existing methods. The network extracts spatio-temporal features through parallel multi-scale convolutional branches and fuses them with an attention mechanism to sharpen transient motor-imagery responses, targeting the non-stationary EEG signals and inter-subject variability that make MI decoding unreliable. The study was published on February 27, 2026, in IEEE Journal of Biomedical and Health Informatics.

The Chinese Clinical Trial Registry (ChiCTR) has logged a study of a brain-computer interface (BCI)-enabled cognitive rehabilitation system, set to test its clinical effect in patients with mild cognitive impairment. Registered on February 25, 2026 as ChiCTR2600119334, the trial is not yet recruiting.

Researchers propose a function predefined convolutional neural network (FPCNN) for decoding spontaneous EEG in brain-computer interfaces. Its learnable function predefined convolution (FPC) layer searches for the key spatial-frequency parameters of spontaneous EEG so the parameters carry clear physical meaning, and it builds trainable orthogonal detectors on the FPC to capture complex phase-varying signals. On three spontaneous EEG datasets, FPCNN outperformed state-of-the-art methods by 2.09%, 3.08% and 3.41%, with single-round training and testing taking just 67.96 and 19.36 seconds on non-GPU hardware, which the authors say makes it suited to EEG processing in diverse environments.

The study presents an EEG emotion recognition framework that couples discriminative channel selection with hierarchical spatial-temporal modeling. Wavelet coherence and mutual information adaptively select informative channels across frequency bands, and a spatial-temporal graph-aware network (STG-Net) models inter-channel spatial relations and the temporal evolution of emotional states before fusing frequency-spatial-temporal features for classification. The authors report better recognition accuracy and model efficiency than state-of-the-art methods.

NeuroXess broke ground on a roughly 14,300-square-meter BCI manufacturing project in Jiangxi province in eastern China. The company said the facility would produce flexible implantable electrodes, neural recording equipment, surgical robots and fully implanted wireless BCI systems, with a target of stable supply at the 10,000-unit scale. NeuroXess expects the factory to begin operating in the second half of 2026, but the capacity and timeline remain company projections.

BrainCo's FocusXin Medical Edition combines a single-channel prefrontal EEG device registered as 浙械注准20232071898 with ADHD rehabilitation software registered as 浙械注准20252211934. The Class II system is intended for supplementary rehabilitation training in children aged 6 to 12 with ADHD. Its defined medical use should not be generalized into approval for an unspecified consumer BCI product.

Tongji Hospital in Wuhan implanted a 64-channel high-throughput flexible electrode array made by Wuhan Neuracom into a 31-year-old man paralysed by a spinal cord injury on 26 November 2025. The procedure was charged under the country's first BCI medical service price schedule, issued by Hubei's healthcare security bureau in March 2025, which sets implantation at 6,552 yuan. China's National Healthcare Security Administration calls it the first BCI operation in the country carried out under a state-set price. This patient's costs were covered by research funding.

The General Office of China's Jiangxi Provincial People's Government issued Several Measures for Accelerating the Development of Brain-Computer Interface Technology and Industry in Jiangxi on November 26, 2025, document number Gan Fu Ting Fa [2025] No. 29, setting out 11 measures across four areas. It was the first provincial document to follow the seven-ministry national guidance and is the only one that sets no quantitative industry targets at all. Its eleventh measure calls for implementing BCI research ethics guidelines and establishing mutual recognition of clinical research ethics review, a first among provincial documents. Other measures deploy a provincial major science and technology programme, back Ganjiang New Area in building a rehabilitation centre and a high-quality EEG dataset, and offer early regulatory engagement plus rewards for Class II and Class III innovative device approvals.

Tongji Hospital in Wuhan announced a brain-computer interface research institute on 5 September 2025, which it says is the first of its kind run by a hospital in China rather than by a university or a company. The institute covers basic research, clinical application and commercialisation, with the stated aim of building a full-chain BCI system spanning prevention, diagnosis, treatment and rehabilitation of neurological disease.

Neuracle Medical Technology (Shanghai) Co., Ltd. ran a clinical trial in China evaluating the safety and efficacy of its implantable wireless brain-computer interface, NEO, in patients with tetraplegia caused by spinal cord injury. The trial is registered on ClinicalTrials.gov as NCT06990412 and is listed as completed; its stated primary purpose was to provide the basis for product registration. NEO is designed to let patients control external devices with brain signals.

Thirteen departments in China's Shandong province, led by the Department of Science and Technology, jointly issued the Action Plan for Scientific and Technological Innovation in the Brain-Computer Interface Industry in Shandong (2025-2027), document number Lu Ke Zi [2025] No. 50, dated May 13, 2025 and released on May 16. By 2027 the province aims for roughly 20 breakthroughs in frontier and core technologies, about 30 innovative technology SMEs, two to three high-level talent teams, and one to two specialised incubators in areas rich in research and education resources, with applications reaching medicine, industry, elderly care, education and entertainment. By 2030 it expects an initial BCI industry ecosystem. The plan organises work into six campaigns and backs Jinan and Qingdao as innovation clusters. It is the first of Shandong's two provincial BCI documents and covers the industrial side.

Eight Sichuan provincial departments in southwestern China led by the Department of Economy and Information Technology issued the Action Plan for Breakthroughs in the Brain-Computer Interface and Human-Machine Interaction Industry in Sichuan (2025-2030) on May 12, 2025. The plan is unusual in making surgical volume a headline metric, targeting 3,000 invasive BCI procedures a year by 2030, along with services for more than 100,000 patients a year with neurodegenerative disease, psychiatric conditions and drug or digital addiction, and rehabilitation device use exceeding 20,000 sessions a year. By 2027 it aims to complete development and medical device registration of three invasive and five non-invasive products, carry out the province's first invasive BCI operation, serve more than 50,000 patients a year, and cultivate 10 chain-leading firms, 100 specialised SMEs and 200 innovative smaller enterprises. Chengdu, Mianyang, Deyang and Nanchong are named as host cities.

The study proposes a discriminative, geometry-aware dimensionality reduction method on the Riemannian manifold for symmetric positive definite (SPD) matrices in EEG classification, aimed at boosting discrimination while reducing information loss. On BCI Competition IV Dataset 1 and Dataset 2a, the method improved classification accuracy by 5.0% and 19.38% respectively, indicating robust performance is retained after dimensionality reduction.

Shanghai's science commission issued a BCI industry action plan centering on medical-grade scenarios, with a clinical timeline for invasive and semi-invasive products.

Three Beijing agencies jointly issued a BCI action plan targeting 3-5 globally influential leaders and around 100 innovative SMEs by 2030.

NeuroXess and Huashan Hospital reported a single-patient test of a 256-channel flexible implantable BCI. In postoperative training, decoding accuracy reached 71% across 142 commonly used Mandarin syllables, with per-character latency below 100 milliseconds. The participant had a language-area tumor and epilepsy, so the company-reported result does not establish performance in people who have lost speech.

Chinese registry ChiCTR2400081380 describes a 96-patient trial of an implantable closed-loop recording and stimulation system for focal drug-resistant epilepsy in people aged 6 to 70. All participants are to receive the implant and then be randomized one month later to active or sham stimulation, 48 per group. The registry provides no evidence that enrollment had passed 50%.